checkpoint-500/trainer_state.json

{
  "best_metric": 0.10345435318998943,
  "best_model_checkpoint": "./whisper-medium-sr-yodas/checkpoint-500",
  "epoch": 0.2442002442002442,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 9.130462646484375,
      "learning_rate": 4.600000000000001e-06,
      "loss": 1.914,
      "step": 25
    },
    {
      "epoch": 0.02,
      "grad_norm": 11.731131553649902,
      "learning_rate": 9.600000000000001e-06,
      "loss": 1.4161,
      "step": 50
    },
    {
      "epoch": 0.04,
      "grad_norm": 7.232382297515869,
      "learning_rate": 9.841379310344828e-06,
      "loss": 1.1372,
      "step": 75
    },
    {
      "epoch": 0.05,
      "grad_norm": 6.721960067749023,
      "learning_rate": 9.66896551724138e-06,
      "loss": 0.8027,
      "step": 100
    },
    {
      "epoch": 0.06,
      "grad_norm": 7.100645065307617,
      "learning_rate": 9.496551724137932e-06,
      "loss": 0.7147,
      "step": 125
    },
    {
      "epoch": 0.07,
      "grad_norm": 9.62185001373291,
      "learning_rate": 9.324137931034484e-06,
      "loss": 0.8326,
      "step": 150
    },
    {
      "epoch": 0.09,
      "grad_norm": 7.819121360778809,
      "learning_rate": 9.151724137931035e-06,
      "loss": 0.7793,
      "step": 175
    },
    {
      "epoch": 0.1,
      "grad_norm": 5.436564922332764,
      "learning_rate": 8.979310344827587e-06,
      "loss": 0.6697,
      "step": 200
    },
    {
      "epoch": 0.11,
      "grad_norm": 6.4688029289245605,
      "learning_rate": 8.806896551724139e-06,
      "loss": 0.6794,
      "step": 225
    },
    {
      "epoch": 0.12,
      "grad_norm": 6.041417121887207,
      "learning_rate": 8.634482758620691e-06,
      "loss": 0.6636,
      "step": 250
    },
    {
      "epoch": 0.13,
      "grad_norm": 9.54858112335205,
      "learning_rate": 8.462068965517241e-06,
      "loss": 0.7643,
      "step": 275
    },
    {
      "epoch": 0.15,
      "grad_norm": 8.129537582397461,
      "learning_rate": 8.289655172413793e-06,
      "loss": 0.6648,
      "step": 300
    },
    {
      "epoch": 0.16,
      "grad_norm": 7.3216872215271,
      "learning_rate": 8.117241379310346e-06,
      "loss": 0.6526,
      "step": 325
    },
    {
      "epoch": 0.17,
      "grad_norm": 6.333176136016846,
      "learning_rate": 7.944827586206898e-06,
      "loss": 0.6882,
      "step": 350
    },
    {
      "epoch": 0.18,
      "grad_norm": 7.942104816436768,
      "learning_rate": 7.77241379310345e-06,
      "loss": 0.6674,
      "step": 375
    },
    {
      "epoch": 0.2,
      "grad_norm": 5.19351863861084,
      "learning_rate": 7.600000000000001e-06,
      "loss": 0.6029,
      "step": 400
    },
    {
      "epoch": 0.21,
      "grad_norm": 5.838120937347412,
      "learning_rate": 7.427586206896552e-06,
      "loss": 0.5979,
      "step": 425
    },
    {
      "epoch": 0.22,
      "grad_norm": 6.667115688323975,
      "learning_rate": 7.255172413793104e-06,
      "loss": 0.6611,
      "step": 450
    },
    {
      "epoch": 0.23,
      "grad_norm": 4.763759613037109,
      "learning_rate": 7.0827586206896555e-06,
      "loss": 0.5857,
      "step": 475
    },
    {
      "epoch": 0.24,
      "grad_norm": 5.6900129318237305,
      "learning_rate": 6.910344827586208e-06,
      "loss": 0.5918,
      "step": 500
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.17614366114139557,
      "eval_runtime": 446.8109,
      "eval_samples_per_second": 3.453,
      "eval_steps_per_second": 0.432,
      "eval_wer": 0.10345435318998943,
      "eval_wer_ortho": 0.20109366731345915,
      "step": 500
    }
  ],
  "logging_steps": 25,
  "max_steps": 1500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 8.16483926016e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}